Best model for dummy explanatory variable

[Roberta Pilgrim]

Hi there!
I'm trying to understand what is the best model to use for this research question: what is the effect of the number of searches, days of use, etc
(x_1, x_2, x _n, etc..) on the probability to become a premium user of a service?

- y is a dummy variable where 0 is NONPREMIUM and 1 is PREMIUM.
- I have cross-sectional data where each id corresponds to a row of data
- The sample counts 80k users/id, they started using the service in different moments and we take time into account with a variable which represents "numbers of days from the activation day"

Which model would you suggest to use?

Thank you so much in advance for your help!
RP

[Carlo Lazzaro]

Roberta:
I would try:

logistic y i.x1 i.x2 i.xn c.number_days##c.number_days